1
$\begingroup$

I have an old HBOT printer (now they call themselves Zmorph).

I had to replace the old hotend. From the looks of it, what I need is a J-head MkV type nozzle (from RepRap wiki).

I've bought and put together the entire hotend but unfortunately when I've tried to heat up the nozzle to 150 °C for a test run, it started to heat up very quickly (compared to the old hotend) and didn't stop at the temperature I've set. I've quickly turned off the heating at around 200 °C, but the temperature still managed to climb up to 280 °C from inertia alone.

Thermistor shows the temperature correctly so its probably not a typical thermal runaway situation. I've measured the resistance on both the old and new heater cartridge and since I've never held a multimeter in my hand before, I need help with interpreting results. Basically it turns out that:

  • old cartridge: resistance=14.5 Ω which means it should be: voltage = 24 V and power = 40 W
  • new cartridge: resistance=6.5 Ω which means it should be: voltage = 12 V and power = 20 W

So, it stands to reason that if I've put a wrong heating cartridge into my hotend I will get:

power = (voltage^2)/resistance = (24^2)/6.5 = 89 W

instead of the correct value I got with the old heating cartridge:

power = (voltage^2)/resistance = (24^2)/14.5 = 40 W

My heating cartridge gets more then double the power it needs (89 W instead of 40 W), so naturally it overheats

A logical solution would be to simply buy a new ceramic heating cartridge with correct parameters.

enter image description here

Does my reasoning make sense?

$\endgroup$
2
  • 2
    $\begingroup$ Yes this makes sense, the current PID schedule is not able to cope with the higher power cartridges. It remains a question if you will get it stable after a PID tuning session. See e.g. this question with this answer, which is very simiar to the behavior you describe. This question can be marked as a duplicate for that question. I was able to get this sort of stable by lowering the power factor of the heating cartridge in firmware. $\endgroup$
    – 0scar
    Commented May 6 at 12:40
  • 2
    $\begingroup$ Thank you, I'm a bit wary of PID tuning on that particular printer. I will try doing it if I have to, but first I'm just going to buy the same type of heating cartridge that was previously mounted and hope it works. $\endgroup$ Commented May 7 at 6:02

1 Answer 1

2
$\begingroup$

Does my reasoning make sense?

Yes this makes sense, the current PID schedule is not able to cope with the higher power cartridges. It remains a question if you will get it stable after a PID tuning session as the heat input is very large.

See e.g. this question with this answer, which is simiar to the behavior you describe.

I have had similar issues when I mixed 24 V and 12 V cartridges in my spares supply and installed the incorrect one by mistake. I was able to get this sort of stable by lowering the power factor of the heating cartridge in firmware, but this is unsupported in the latest version of Marlin. You might have better luck with the "Model Predictive Control for hotend" model. Alternatively, and the best solution (I ended up doing as well) is inserting the correct heater cartridge.

$\endgroup$
3
  • $\begingroup$ Thank you, I've got a correct heater cartridge today. After mounting it in the hotend the temperature still climbed past the set level e.g. when I've set the temperature to 150 it climbed to 230 before dropping, but PID loop at least got it under control. $\endgroup$ Commented May 10 at 12:04
  • $\begingroup$ There is no option to autotune PID in the printer menu and there is no way of hooking it up to a laptop. I know very little about printer software in general, but it seems that software is precompiled on the same SD card on which I store gcode files because vscode opens it as a binary file. The printer screen is completely blank and the lights rapidly turn on and off when I try to turn the printer on without an SD card inserted. SD card also contains a config file with PID coefficients, so through trial and error I've managed to get the fluctuations to ±10 °C. I will try again come monday. $\endgroup$ Commented May 10 at 12:15
  • $\begingroup$ @Piotrekdoro Can you look up the controller board make and model? Based on the board you could probably update the board through SD card. Is there no interface port on the board? E.g. ISB? $\endgroup$
    – 0scar
    Commented May 10 at 17:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .